pubmed.ncbi.nlm.nih.gov

Extending the Spectrum of Dysgraphia: A Data Driven Strategy to Estimate Handwriting Quality - PubMed

  • ️Wed Jan 01 2020

Extending the Spectrum of Dysgraphia: A Data Driven Strategy to Estimate Handwriting Quality

Thibault Asselborn et al. Sci Rep. 2020.

Abstract

This paper proposes new ways to assess handwriting, a critical skill in any child's school journey. Traditionally, a pen and paper test called the BHK test (Concise Evaluation Scale for Children's Handwriting) is used to assess children's handwriting in French-speaking countries. Any child with a BHK score above a certain threshold is diagnosed as 'dysgraphic', meaning that they are then eligible for financial coverage for therapeutic support. We previously developed a version of the BHK for tablet computers which provides rich data on the dynamics of writing (acceleration, pressure, and so forth). The underlying model was trained on dysgraphic and non-dysgraphic children. In this contribution, we deviate from the original BHK for three reasons. First, in this instance, we are interested not in a binary output but rather a scale of handwriting difficulties, from the lightest cases to the most severe. Therefore, we wish to compute how far a child's score is from the average score of children of the same age and gender. Second, our model analyses dynamic features that are not accessible on paper; hence, the BHK is useful in this instance. Using the PCA (Principal Component Analysis) reduced the set of 53 handwriting features to three dimensions that are independent of the BHK. Nonetheless, we double-checked that, when clustering our data set along any of these three axes, we accurately detected dysgraphic children. Third, dysgraphia is an umbrella concept that embraces a broad variety of handwriting difficulties. Two children with the same global score can have totally different types of handwriting difficulties. For instance, one child could apply uneven pen pressure while another one could have trouble controlling their writing speed. Our new test not only provides a global score, but it also includes four specific score for kinematics, pressure, pen tilt and static features (letter shape). Replacing a global score with a more detailed profile enables the selection of remediation games that are very specific to each profile.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Figure 1
Figure 1

The value of the feature: Number of Peaks in Velocity per second (#15) as a function of age is plotted for all individuals in our database. The red points represent children recruited in schools, while the green points represent children with dysgraphia. The interpolated function representing the mean (fmean) is plotted as well as the interpolated function representing the standard deviation (fstd). Notice that the points for the children with dysgraphia seem to be located further from the mean compared to the points for their peers.

Figure 2
Figure 2

The amount of variability explained by the three first axes of the PCA. The absolute feature importance for each of the three first axes is also represented (in the radar chart) and was sorted according to the four categories. For reasons of clarity, the graphs were plotted with only the six most important features for each category (static, kinematic, tilt and pressure). The features are represented by numbers. The correspondence between the numbers and the feature names and descriptions can be found in the Method section.

Figure 3
Figure 3

K-Means Algorithm applied to our three-dimensional dataset (defined by the features projected on the three first axes of the PCA), with two clusters under three different projections. The crosses represent children with severe handwriting difficulties (as determined by their BHK scores), whereas the points represent children recruited in schools. The colors represent the two different clusters.

Figure 4
Figure 4

Scores computed for all the children in our database. The red points represent children recruited from schools, while the green points represent children with dysgraphia. Four threshold values (very severe, severe, moderate, light) were used to divide handwriting difficulties into five categories with 2% of school children below the very severe threshold (this allowed us to compute the vs value), 8.6% of school children below the severe threshold (this allowed us to compute the s value, which is the current dysgraphia threshold), 15% of children below the moderate threshold (this allowed us to compute the m value) and 25% of children below the light threshold (this allowed us to compute the l value). It is important to note that these thresholds have been set as examples, meaning that any other values can be used. The score is proportional to the handwriting quality.

Figure 5
Figure 5

Different handwriting profiles of children with severe handwriting difficulties. We can see that the same condition (dysgraphia) can be expressed in very different ways, with children presenting difficulties in different and independent skills (Kinematic, Pressure, Tilt and Static).

Figure 6
Figure 6

Illustration of the Handwriting density. The space is split into 20-pixel-wide square cells. Using the number of points per cell, we can then compute a density.

Figure 7
Figure 7

The whole process used to extract the frequency spectrum of our signal. (a) We first divided the BHK text into bins of 600 points. (b) For each packet, the signal was extracted. (c) We computed the Fourier transform of the signal. (d) We took the average of all signals and performed a normalization. (e) In these sample signals extracted from the data, the red dots are the point coordinates recorded by the device during handwriting, the vectors in blue are “local” vectors linking two consecutive points and the vector in green is the "global” vector (average of the nine blue vectors) representing the global direction of the handwriting. The cross product of these two vectors gives us an indication of the smoothness/shakiness of the handwriting. The right side of the figure comes from a writer with smoother/less shaky handwriting than the writer producing the one on the left, and the cross product operation will detect this difference. This image has been adapted from.

Figure 8
Figure 8

The two angles (Tilt-Azimuth and Tilt-Altitude) recorded for the pen.

Similar articles

Cited by

References

    1. Graham S. The role of production factors in learning disabled students’ compositions. J. educational psychology. 1990;82:781. doi: 10.1037/0022-0663.82.4.781. - DOI
    1. Berninger VW, et al. Treatment of handwriting problems in beginning writers: Transfer from handwriting to composition. J. Educ. Psychol. 1997;89:652. doi: 10.1037/0022-0663.89.4.652. - DOI
    1. Bourdin B, Fayol M. Is written language production more difficult than oral language production? a working memory approach. Int. journal psychology. 1994;29:591–620. doi: 10.1080/00207599408248175. - DOI
    1. Bourdin B, Fayol M. Is graphic activity cognitively costly? A developmental approach. Read. & Writ. 2000;13:183–196. doi: 10.1023/A:1026458102685. - DOI
    1. Feder K, Majnemer A. Handwriting development, competency, and intervention. Dev. Medicine Child Neurol. 2007;49:312–317. doi: 10.1111/j.1469-8749.2007.00312.x. - DOI - PubMed

Publication types

MeSH terms